By arbiterhq
Gemini-powered image intelligence (describe, generate, edit, ocr, detect, extract, diff, compare, analyze, batch) plus Sharp-powered local manipulation (info, resize, upscale, crop, palette, optimize, convert, sprite sheets).
Cross-platform agent skills for Claude Code, Codex CLI, and Gemini CLI. By Posthuman Resources LLC.
Browser automation wrapping Vercel's agent-browser CLI. Navigate pages, fill forms, click buttons, take screenshots, scrape structured data, and automate browser workflows.
Image generation via Nano Banana (Google Gemini image models) and local image processing via Sharp, ImageMagick, and FFmpeg. Generate from prompts, edit existing images, create transparent assets, resize, convert, and batch process.
Opinionated git workflow management. Atomic one-idea-per-commit methodology, worktrees for parallel development, change stacking across dependent branches, GitHub PR workflows, and merge strategies.
/plugin marketplace add arbiterhq/agent-skills
Then install individual plugins:
/plugin install browser-buddy@arbiterhq
/plugin install artistic-vision@arbiterhq
/plugin install git-ideas@arbiterhq
Using the built-in skill installer:
$skill-installer install browser-buddy from arbiterhq/agent-skills
Or manually:
git clone https://github.com/arbiterhq/agent-skills.git
cd agent-skills && bash install.sh
gemini extensions install https://github.com/arbiterhq/agent-skills
Works with Claude Code, Codex, Gemini CLI, Cursor, and 40+ other agents:
npx skills add arbiterhq/agent-skills
This repo is a Bun workspace. To get started:
bun install
Each plugin under plugins/* is a workspace member. The root package is private and not published to npm; distribution happens via the Git repo itself (see Installation above).
agent-skills/
package.json # Bun workspace root (private)
.claude-plugin/marketplace.json # Claude Code marketplace manifest
gemini-extension.json # Gemini CLI extension manifest
plugins/ # Canonical skill sources (Claude Code plugins)
browser-buddy/
package.json # Workspace member
.claude-plugin/plugin.json # Claude Code plugin manifest
skills/browser-buddy/SKILL.md
artistic-vision/
git-ideas/
skills/ # Symlinks for Codex and Gemini discovery
CLAUDE.md # Claude Code project context
AGENTS.md # Codex project context
GEMINI.md # Gemini CLI project context
Skills live in plugins/<name>/skills/<name>/SKILL.md with the canonical source of truth. The skills/ directory at the root contains symlinks for cross-tool compatibility.
To add a new skill:
plugins/<skill-name>/.claude-plugin/plugin.json manifestskills/<skill-name>/SKILL.md with YAML frontmatterskills/ directory.claude-plugin/marketplace.jsonMIT - Posthuman Resources LLC
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub arbiterhq/agent-skills --plugin artistic-visionBrowser automation via Vercel's agent-browser CLI. Drives real Chrome for navigation, form filling, scraping, screenshotting, video recording, and web app testing. Built around the snapshot-ref interaction pattern for token-efficient agent use.
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Frontend design skill for UI/UX implementation
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Memory compression system for Claude Code - persist context across sessions
Marketing skills for AI agents — conversion optimization, copywriting, SEO, paid ads, ad creative, and growth
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.